search

quark0 / darts

Differentiable architecture search for convolutional and recurrent networks
https://arxiv.org/abs/1806.09055
Apache License 2.0
3.92k stars 843 forks source link

Loading the model after training #173

Open Tahmid1406 opened 2 months ago

Tahmid1406 commented 2 months ago

After the search process, ("using the train_search") I end up with a weights.pt file, a log file, reduction and normal cells.

My question is how do I load the best model and do inference on it for the cifar10 model?

imanehmz commented 2 weeks ago

you should add the resulting architecture in the genotypes file, and have something like : MYDARTS = genotype(....), then you can use the train.py file to train the resulting architecture, use the argument --arch=MYDARTS